This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [C++ PATCH] Don't create an INTEGER_CST for aggregates (emptystructs)
- From: Paolo Carlini <pcarlini at suse dot de>
- To: Jason Merrill <jason at redhat dot com>
- Cc: Gabriel Dos Reis <gdr at integrable-solutions dot net>,Andrew Pinski <pinskia at physics dot uc dot edu>,GCC Patches <gcc-patches at gcc dot gnu dot org>,libstdc++ <libstdc++ at gcc dot gnu dot org>
- Date: Sun, 13 Mar 2005 20:36:52 +0100
- Subject: Re: [C++ PATCH] Don't create an INTEGER_CST for aggregates (emptystructs)
- References: <9a23c41857b1b1161e9d02c278ba5592@physics.uc.edu> <42308EA6.1010408@suse.de> <85c98d4cd9d0e7435f8517e9a69ac136@physics.uc.edu> <m3ll8ty00m.fsf@uniton.integrable-solutions.net> <4232B7B4.4040207@suse.de> <m364zxgnvm.fsf@uniton.integrable-solutions.net> <xypy8cruz72.fsf@miranda.boston.redhat.com>
Hi Jason,
On 12 Mar 2005 11:27:41 +0100, Gabriel Dos Reis <gdr@integrable-solutions.net> wrote:
Second, the case of stack-based CPUs, e.g. x86-like.
The caller does not need any push of such trailing empty class
arguments on stack -- and the callee knows it has no business trying
to pop such value. Consequently adjustments need to be done for the
stack frame "depth".
This assumes that we control both caller and callee, which violates the
idea of a standardized ABI.
Ok. Therefore, which are the implications of this? I mean, is it
unavoidable that
the code generated on powerpc is better that on x86 (for the reasons
explained by
Andrew in the audit trail)? Sorry about the direct question.
Thanks,
Paolo.